Wind Projects - Initial Site Selection

The initial site selection is the first phase in the development of any wind energy project. In this phase appropriate sites should be identified and their wind potentials should be estimated. By identifying environmental, technical, commercial and political constraints of the sites the project developer can decide whether a more extensive feasibility study should be conducted. As a starting point many developers visit the possible project sites, gathering first impressions about topography and infrastructure (roads, dwellings, grid-connection). A central task in this phase, available environmental and technical data must be collected.

First estimation of energy yields and related data requirements

The data requirements for a first energy-yield estimation of a wind project are defined by the following function

$E=\sum _{i=1}^{i=1}H(U_{i})P(U_{i})$

where H is the number of hours with windspeed U_i and P is the value of the power curve for the proposed wind turbine for the windspeed U_i.

A wind speed distribution function provides information about the frequency of different wind velocities occuring during a year. The generation of these distribution function requires long-time (for minimum one year) measurement. Thus for initial estimation, regional available data should be used, providing only rough information but keeping the initial costs low. The so-called power distribution function (PDF) can be obtained from a wind atlas (for example the NREL Oaxaca Wind Energy Atlas provides regional wind data for a state of Mexico). The US National Renewable Energy Laboratory (NREL) publishes information about wind speed distributions for many regions of the world on its Website[[3]]. PDFs are based on Weibull-Distributions and take into account local conditions like the roughness of the surrounding terrain (given by surface characteristics, obstacles and topology) are calculated for any wind direction using a divisions of 12 30° sectors. To use the values of the PDF in the function presented above, wind velocity values of all directions are integrated.

The power curve of a wind turbine depicts the energy output of the turbine at different wind speeds U_i. Paul Gipe suggests that the reliability of power curves especially for small wind turbines is limited, but for the initial estimation the available power curves are sufficient.

In case data availability is very limited (for instance only from remote weather stations), wind velocity modelling including meteorological data and site characteristics can be used.